{"id":"https://openalex.org/W4381328970","doi":"https://doi.org/10.1145/3589273","title":"Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines","display_name":"Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines","publication_year":2023,"publication_date":"2023-06-13","ids":{"openalex":"https://openalex.org/W4381328970","doi":"https://doi.org/10.1145/3589273"},"language":"en","primary_location":{"id":"doi:10.1145/3589273","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3589273","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3589273","source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3589273","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081322278","display_name":"Stefan Grafberger","orcid":"https://orcid.org/0000-0002-9884-9517"},"institutions":[{"id":"https://openalex.org/I4210135670","display_name":"Amsterdam University of the Arts","ror":"https://ror.org/04dde1554","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210135670"]},{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Stefan Grafberger","raw_affiliation_strings":["University of Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0002-9884-9517","affiliations":[{"raw_affiliation_string":"University of Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I4210135670","https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034924491","display_name":"Paul Groth","orcid":"https://orcid.org/0000-0003-0183-6910"},"institutions":[{"id":"https://openalex.org/I4210135670","display_name":"Amsterdam University of the Arts","ror":"https://ror.org/04dde1554","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210135670"]},{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Paul Groth","raw_affiliation_strings":["University of Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0003-0183-6910","affiliations":[{"raw_affiliation_string":"University of Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I4210135670","https://openalex.org/I887064364"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5090934117","display_name":"Sebastian Schelter","orcid":"https://orcid.org/0000-0003-4722-5840"},"institutions":[{"id":"https://openalex.org/I4210135670","display_name":"Amsterdam University of the Arts","ror":"https://ror.org/04dde1554","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210135670"]},{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Sebastian Schelter","raw_affiliation_strings":["University of Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0003-4722-5840","affiliations":[{"raw_affiliation_string":"University of Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I4210135670","https://openalex.org/I887064364"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.8553,"has_fulltext":true,"cited_by_count":18,"citation_normalized_percentile":{"value":0.92024603,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":"1","issue":"2","first_page":"1","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.9990000128746033,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.9990000128746033,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10260","display_name":"Software Engineering Research","score":0.9912999868392944,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9805999994277954,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8034433126449585},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.8015333414077759},{"id":"https://openalex.org/keywords/pipeline-transport","display_name":"Pipeline transport","score":0.772910475730896},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.5538804531097412},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.5503168702125549},{"id":"https://openalex.org/keywords/data-pre-processing","display_name":"Data pre-processing","score":0.4826396107673645},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.46390339732170105},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3560669720172882},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.28386300802230835},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.1701415777206421},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10799163579940796}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8034433126449585},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.8015333414077759},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.772910475730896},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5538804531097412},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.5503168702125549},{"id":"https://openalex.org/C10551718","wikidata":"https://www.wikidata.org/wiki/Q5227332","display_name":"Data pre-processing","level":2,"score":0.4826396107673645},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.46390339732170105},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3560669720172882},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28386300802230835},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.1701415777206421},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10799163579940796},{"id":"https://openalex.org/C87717796","wikidata":"https://www.wikidata.org/wiki/Q146326","display_name":"Environmental engineering","level":1,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3589273","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3589273","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3589273","source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},{"id":"pmh:oai:dare.uva.nl:publications/100f1a2d-a34e-4558-8d31-2664ddaa8f5e","is_oa":true,"landing_page_url":"https://handle.uba.uva.nl/personal/pure/en/publications/automating-and-optimizing-datacentric-whatif-analyses-on-native-machine-learning-pipelines(100f1a2d-a34e-4558-8d31-2664ddaa8f5e).html","pdf_url":"https://pure.uva.nl/ws/files/169536235/Automating_and_Optimizing_Data-Centric_What-If_Analyses_on_Native_Machine_Learning_Pipelines.pdf","source":{"id":"https://openalex.org/S4306400088","display_name":"UvA-DARE (University of Amsterdam)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887064364","host_organization_name":"University of Amsterdam","host_organization_lineage":["https://openalex.org/I887064364"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Grafberger, S, Groth, P & Schelter, S 2023, 'Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines', Proceedings of the ACM on Management of Data, vol. 1, no. 2, 128. https://doi.org/10.1145/3589273","raw_type":"info:eu-repo/semantics/article"},{"id":"pmh:oai:dare.uva.nl:openaire_cris_publications/100f1a2d-a34e-4558-8d31-2664ddaa8f5e","is_oa":true,"landing_page_url":"https://hdl.handle.net/11245.1/100f1a2d-a34e-4558-8d31-2664ddaa8f5e","pdf_url":null,"source":{"id":"https://openalex.org/S4306400088","display_name":"UvA-DARE (University of Amsterdam)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887064364","host_organization_name":"University of Amsterdam","host_organization_lineage":["https://openalex.org/I887064364"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Grafberger, S, Groth, P & Schelter, S 2023, 'Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines', Proceedings of the ACM on Management of Data, vol. 1, no. 2, 128. https://doi.org/10.1145/3589273","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1145/3589273","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3589273","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3589273","source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.5,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4381328970.pdf","grobid_xml":"https://content.openalex.org/works/W4381328970.grobid-xml"},"referenced_works_count":48,"referenced_works":["https://openalex.org/W1987605824","https://openalex.org/W2047809518","https://openalex.org/W2069400400","https://openalex.org/W2098416578","https://openalex.org/W2295598076","https://openalex.org/W2411006959","https://openalex.org/W2547386789","https://openalex.org/W2548122763","https://openalex.org/W2780624097","https://openalex.org/W2807799957","https://openalex.org/W2904201207","https://openalex.org/W2905588001","https://openalex.org/W2926314329","https://openalex.org/W2929941791","https://openalex.org/W2935765604","https://openalex.org/W2949054050","https://openalex.org/W2956281901","https://openalex.org/W2969946509","https://openalex.org/W2970641574","https://openalex.org/W2971290973","https://openalex.org/W2971428651","https://openalex.org/W2988175183","https://openalex.org/W2989755645","https://openalex.org/W3093752251","https://openalex.org/W3105977086","https://openalex.org/W3139291533","https://openalex.org/W3166319166","https://openalex.org/W3173683037","https://openalex.org/W3194157648","https://openalex.org/W4210354100","https://openalex.org/W4221151137","https://openalex.org/W4225853030","https://openalex.org/W4226303346","https://openalex.org/W4281386293","https://openalex.org/W4281626245","https://openalex.org/W4282038795","https://openalex.org/W4282570934","https://openalex.org/W4312992260","https://openalex.org/W6614148910","https://openalex.org/W6675354045","https://openalex.org/W6686224169","https://openalex.org/W6763088532","https://openalex.org/W6772203414","https://openalex.org/W6783187084","https://openalex.org/W6792494324","https://openalex.org/W6799868450","https://openalex.org/W6839002650","https://openalex.org/W6910653033"],"related_works":["https://openalex.org/W4380433113","https://openalex.org/W4386072068","https://openalex.org/W252339960","https://openalex.org/W2390529043","https://openalex.org/W2378320433","https://openalex.org/W2989490741","https://openalex.org/W2358343511","https://openalex.org/W2071821326","https://openalex.org/W2051877971","https://openalex.org/W1970117064"],"abstract_inverted_index":{"Software":[0],"systems":[1,33],"that":[2,20,262,308],"learn":[3],"from":[4],"data":[5,24,46,54,71,164,282],"with":[6,37,68,220,253,267,287],"machine":[7],"learning":[8],"(ML)":[9],"are":[10,34,127],"used":[11],"in":[12,31,247,273,318],"critical":[13],"decision-making":[14],"processes.":[15],"Unfortunately,":[16],"real-world":[17],"experience":[18],"shows":[19],"the":[21,60,106,109,182,195,215,231,268,280,309,314,319,324],"pipelines":[22,121,245],"for":[23,155,171,212],"preparation,":[25],"feature":[26],"encoding":[27],"and":[28,62,98,135,152,175,180,198,235,260,276,306,316],"model":[29,315],"training":[30],"ML":[32,87,120,160,173,244],"often":[35],"brittle":[36],"respect":[38,69],"to":[39,49,58,70,103,119,129,142,148,166,176,191,194,229,249,279,300,303,313],"their":[40,136],"input":[41,281],"data.":[42],"As":[43],"a":[44,80,90,95,201,209],"consequence,":[45],"scientists":[47,165],"have":[48],"run":[50],"different":[51],"kinds":[52],"of":[53,64,115,200,237,270,298,328],"centric":[55],"what-if":[56,77,157,169],"analyses":[57,78,158,170,289],"evaluate":[59,258],"robustness":[61],"reliability":[63],"such":[65],"pipelines,":[66,294],"e.g.,":[67],"errors":[72],"or":[73],"preprocessing":[74],"techniques.":[75],"These":[76],"follow":[79],"common":[81],"pattern:":[82],"they":[83,126],"take":[84],"an":[85,172],"existing":[86,116,132],"pipeline,":[88,174],"create":[89],"pipeline":[91,101,133,184,189,217,232,271],"variant":[92,102,233],"by":[93],"introducing":[94],"small":[96],"change,":[97],"execute":[99,181],"this":[100],"see":[104],"how":[105,228],"change":[107],"impacts":[108],"pipeline's":[110],"output":[111],"score.":[112],"The":[113],"application":[114],"analysis":[117],"techniques":[118],"is":[122,277,311],"technically":[123],"challenging":[124],"as":[125],"hard":[128],"integrate":[130],"into":[131],"code":[134],"execution":[137],"introduces":[138],"large":[139],"overheads":[140],"due":[141],"repeated":[143],"work.":[144],"We":[145,256],"propose":[146],"mlwhatif":[147,162],"address":[149],"these":[150,205],"integration":[151],"efficiency":[153],"challenges":[154],"data-centric":[156],"on":[159,204,290],"pipelines.":[161],"enables":[163],"declaratively":[167],"specify":[168,192],"automatically":[177],"generate,":[178],"optimize":[179],"required":[183],"variants.":[185],"Our":[186],"approach":[187],"employs":[188],"patches":[190],"changes":[193],"data,":[196],"operators":[197],"models":[199],"pipeline.":[202,320],"Based":[203],"patches,":[206],"we":[207,226,241,295,322],"define":[208],"multi-query":[210],"optimizer":[211,236],"efficiently":[213],"executing":[214],"resulting":[216],"variants":[218,272],"jointly,":[219],"four":[221,288],"subsumption-based":[222],"optimization":[223],"rules.":[224],"Subsequently,":[225],"detail":[227],"implement":[230],"generation":[234],"mlwhatif.":[238,329],"For":[239],"that,":[240],"instrument":[242],"native":[243],"written":[246],"Python":[248],"extract":[250],"dataflow":[251],"plans":[252],"re-executable":[254],"operators.":[255],"experimentally":[257],"mlwhatif,":[259],"find":[261,307],"its":[263],"speedup":[264,310],"scales":[265],"linearly":[266],"number":[269],"applicable":[274],"cases,":[275],"invariant":[278,312],"size.":[283],"In":[284],"end-to-end":[285],"experiments":[286],"more":[291],"than":[292],"60":[293],"show":[296],"speedups":[297],"up":[299],"13x":[301],"compared":[302],"sequential":[304],"execution,":[305],"featurization":[317],"Furthermore,":[321],"confirm":[323],"low":[325],"instrumentation":[326],"overhead":[327]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}